How your computer can find a website by its name. – DeGreekifying Technology #1

in Uncategorized

This is the first in a series of posts designed to help explain how computers and the Internet work in plain English. This series was inspired by our friend Mercedes M. Yardley.

Have you ever wondered how your web browser can find a website like Google automatically?

You can’t just key a name into your phone if they’re not a contact, your TV won’t find the latest episode of “Bones” automatically. (Yes, your Tivo/DVR might but they’re computers too.) How can your computer do it?

The answer is a near-magical background service called “DNS” (Domain Name Service). Its purpose is to save you from having to remember that is really That number is called an “IP address.” (I’ll explain IP Addresses fully in a future post.)

So DNS is a Rolodex for keeping track of computers. Yes, sometimes us computer geeks do make things easier for everyone.

When DNS is broken, you’ll see an error similar to this:

DNS is elegantly simple in how it works.

Every domain (,, etc.) declares a few servers that definitively know the names and numbers for all the computers that belong to that domain. These are “authoritative name servers”. An authoritative name server can handle DNS for many domains.

A bunch of large Internet companies manage special servers to keep track of every domain and their respective authoritative name servers. These are “root name servers”.

For each root name server, there are thousands of authoritative name servers. Imagine the root name servers as the point of a pyramid and the authoritative name servers as the pyramid’s base.

That’s pretty much how the server side works. On to the client (your!) side.

Your ISP (usually your phone or cable company) manages servers for their clients to use for doing DNS lookups. Don’t worry, it’s a lot of trouble to try and track what sites you visit. Yes, your router (that device that connects your computers to the internet) might also run a DNS lookup server and if so, it would go between Steps 2 and 3 below.

Let’s go over what happens when you enter the name of a web site in your browser.

Step 1) You enter “” into your web browser.
Step 2) Your browser asks your computer if it knows our IP address. If so, you’re off and surfing.
Step 3) If not, your computer asks your ISP if they know it.
Step 4) If not, the ISP’s name server asks any root name server the name of our authoritative name server. It then gets the info from our authoritative name server and answers your computer’s query. In case you ask again, your ISP will save the info for a while.

So instead of everyone needing ginormous servers to keep track of every device on the Internet, this system allows for lots of little servers to pass each other only the information needed. The only big servers needed are the root name servers.

So what do you think, was that sufficiently educational without being too confusing? Your feedback will guide the depth of these articles.

And if there is anything you’d like us to write about, let us know.


Non-Techie Author October 7, 2010

A giant Rolodex? Hmm, I like that. Except – and here’s the author part of me – who runs these root servers? Who pays to keep them going? Who pays for their internet connection? Is this a government thing? Or do the phone companies do this out of the goodness of their hearts? When we hear about the government shutting down a terrorist website, do they mess with that website’s entry in the Rolodex somehow – redirecting all the requests to go there to a uh-oh page instead? What would happen if some bad guy messed with these?

You know us authors, we’re all about the “what if”. πŸ™‚

Mercedes October 7, 2010

Successfully deGreekified! πŸ˜€

Masonian October 9, 2010

Hey! I got all that! Pretty cool.

Wow, I feel all geekified. I shall retire to the nerdery now and play WoW.


{ 1 trackback }

Previous post:

Next post: